FCPXML subtitles for Final Cut Pro and DaVinci Resolve

FCPXML is the file format that turns a transcript into real title clips on an editing timeline, instead of a flat caption track. This page covers what it is, every caption look you can export as FCPXML, and how Final Cut Pro and DaVinci Resolve each import it. Below is the preview of those looks, playing.

What an FCPXML file is

FCPXML is Apple's interchange format for Final Cut Pro: an XML description of a timeline, with clips, lanes, titles and their styling, all placed at exact frame positions. A subtitle file such as SRT is a list of text cues with start and end times. An FCPXML is a small edit, so when you import it you get objects your editor already knows how to handle.

For captions, that means each line arrives as a title clip on a lane above your video. You can retype it, restyle it, move it, keyframe it or delete it, exactly as if you had made it by hand.

FCPXML or SRT

• Use SRT when you want a plain caption track, a sidecar file for YouTube or a player, or captions any application can read.
• Use FCPXML when the captions are part of the picture: styled, positioned, animated, or one word at a time. None of that fits in an SRT.
• Both come from the same transcript and land on the same frames, so you can download both and decide on the timeline.

FCPXML is not Premiere XML

Premiere Pro cannot read FCPXML. It reads the older Final Cut Pro 7 XML (xmeml), which shares an ancestor and a name and very little else. Handing Premiere an .fcpxml file gets an error rather than a timeline. For Premiere there is a marker XML that puts a named marker at the start of every caption line, alongside the SRT; see how to add subtitles in Premiere Pro for that route.

Every look you can export as FCPXML

These are all ordinary FCPXML files. None needs a plugin, a Motion template you have to install or a paid title pack.

Plain titles

One editable title per caption line, with your font, colour, outline and placement. Final Cut reads it as Basic Titles, and Resolve imports it as Text+ titles, which it has no other way to make in bulk.

Karaoke highlight

The whole line holds on screen and the spoken word lights up. The file has one title per word, each carrying the line in up to three colour runs. Works in both editors.

Word reveal

The line starts empty and each word appears as it is spoken and stays, so the line builds itself. The words already shown never shift as the rest arrive. Built for Final Cut; check the first few lines in Resolve before a long edit.

Typewriter

Each line types itself out on Final Cut's own Typewriter title, timed so the typing finishes as the last word is spoken. One title per line. Built for Final Cut; check it in Resolve first.

Word build

Each line builds word by word on one of Final Cut's own Build In/Out titles, with a choice of how words arrive and whether the finished line then moves. Final Cut only: Resolve does not run these titles.

Frame rate decides whether it lines up

Speech recognition returns times in seconds. A timeline counts frames. Every time in these files is written as an exact number of frames at the rate you pick (23.976, 24, 25, 29.97, 30, 50, 59.94, 60), so a matching rate lands each title on its frame and a mismatched one drifts further the longer the video runs. Pick the rate of the timeline you are importing into, which is not always the rate the footage was shot at.

DaVinci Resolve reads it as Text+

Resolve's subtitle track cannot be styled like a title, and Resolve has no command that converts one into Text+. Importing an FCPXML as a timeline is the one-step route: every caption becomes a Text+ title, which is a Fusion node and takes any styling Fusion can do. Style one, then copy it and use Paste Attributes on the rest.

When the import goes wrong

• Resolve shows an error: the file was opened with Import → Subtitle. Use File → Import → Timeline.
• Premiere refuses the file: Premiere cannot read FCPXML. Use the Premiere marker XML or the SRT.
• Titles start on time and drift late: the frame rate picked does not match the timeline. Export again at the timeline's rate.

Already have an SRT, VTT or SBV file? The converter writes an FCPXML from it in your browser, with no transcription needed.

Frequently asked questions